def test_runtime():
def normalized_random_vector(length):
v=numpy.random.randint(1,10,length)
sum_=v.sum()+10
return numpy.array(v/float(sum_))
len_distributions = 2500
len_q = 2500
distributions=[normalized_random_vector(len_q) for i in range(len_distributions)]
combine_dempsters_rule(distributions)
def test_runtime():
def normalized_random_vector(length):
v = numpy.random.randint(1, 10, length)
sum_ = v.sum() + 10
return numpy.array(v / float(sum_))
len_distributions = 2500
len_q = 2500
distributions = [
normalized_random_vector(len_q) for i in range(len_distributions)
]
combine_dempsters_rule(distributions)
def test_simple_mass_distribution():
distributions = [numpy.array([0.3, 0.4, 0.2]), numpy.array([0.1, 0.1, 0.4])]
expected_combined = numpy.array([ 0.16, 0.21, 0.2 ])
expected_conflict = 0.39
expected_theta = 0.04
combined, conflict, theta = combine_dempsters_rule(distributions)
assert_almost_equal(combined, expected_combined)
assert_almost_equal(conflict, expected_conflict)
assert_almost_equal(theta, expected_theta)
def test_simple_mass_distribution():
distributions = [
numpy.array([0.3, 0.4, 0.2]),
numpy.array([0.1, 0.1, 0.4])
]
expected_combined = numpy.array([0.16, 0.21, 0.2])
expected_conflict = 0.39
expected_theta = 0.04
combined, conflict, theta = combine_dempsters_rule(distributions)
assert_almost_equal(combined, expected_combined)
assert_almost_equal(conflict, expected_conflict)
assert_almost_equal(theta, expected_theta)
